• Journal of the Korean Society for Precision Engineering
• /
• v.28 no.5
• /
• pp.606-613
• /
• 2011

In this paper, patterns of laser scattering and detection of micro defects have been investigated based on Rayleigh criterion for silicon wafer in solar cell. Also, a new laser scattering mechanism is designed using characteristics of light scattering against silicon wafer surfaces. Its parameters are to be optimally selected to obtain effective and featured patterns of laser scattering. The optimal parametric ranges of laser scattering are determined using the mean intensity of laser scattering. Scattering patterns of micro defects are investigated at the extracted parameter region. Among a lot of pattern features, both maximum connected area and number of connected component in patterns of laser scattering are regarded as the important information for detecting micro defects. Their usefulness is verified in the experiment.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.35 no.5
• /
• pp.341-348
• /
• 2015

Active infrared thermography methods have been known to possess good fault detection capabilities for the detection of defects in materials compared to the conventional passive thermal infrared imaging techniques. However, the reliability of the technique has been under scrutiny. This paper proposes the lock-in thermography technique for the detection and estimation of artificial subsurface defect size and depth with uncertainty measurement.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.1
• /
• pp.245-265
• /
• 2022

In response to problems such as insufficient extraction information, low detection accuracy, and frequent misdetection in the field of Thangka image defects, this paper proposes a YOLOv5 prediction algorithm fused with the attention mechanism. Firstly, the Backbone network is used for feature extraction, and the attention mechanism is fused to represent different features, so that the network can fully extract the texture and semantic features of the defect area. The extracted features are then weighted and fused, so as to reduce the loss of information. Next, the weighted fused features are transferred to the Neck network, the semantic features and texture features of different layers are fused by FPN, and the defect target is located more accurately by PAN. In the detection network, the CIOU loss function is used to replace the GIOU loss function to locate the image defect area quickly and accurately, generate the bounding box, and predict the defect category. The results show that compared with the original network, YOLOv5-SE and YOLOv5-CBAM achieve an improvement of 8.95% and 12.87% in detection accuracy respectively. The improved networks can identify the location and category of defects more accurately, and greatly improve the accuracy of defect detection of Thangka images.

• International Journal of Internet, Broadcasting and Communication
• /
• v.14 no.2
• /
• pp.136-146
• /
• 2022

Some surface defects make the weak quality of steel materials. To limit these defects, we advocate a one-stage detector model RetinaNet among diverse detection algorithms in deep learning. There are several backbones in the RetinaNet model. We acknowledged two backbones, which are ResNet50 and VGG19. To validate our model, we compared and analyzed several traditional models, one-stage models like YOLO and SSD models and two-stage models like Faster-RCNN, EDDN, and Xception models, with simulations based on steel individual classes. We also performed the correlation of the time factor between one-stage and two-stage models. Comparative analysis shows that the proposed model achieves excellent results on the dataset of the Northeastern University surface defect detection dataset. We would like to work on different backbones to check the efficiency of the model for real world, increasing the datasets through augmentation and focus on improving our limitation.

• Proceedings of the KSR Conference
• /
• 2011.05a
• /
• pp.548-553
• /
• 2011

The lock-in thermography was employed to evaluate the defects in railway bogies. Prior to the actual application on railway bogies, in order to assess the detectability of known flaws, the calibration reference panel was prepared with various dimensions of artificial flaws. The panel was composed of polymer matrix composites, which were the same material with actual bogies. Through lock-in thermography evaluation, the optimal frequency of heat source was determined for the best flaw detection. Based on the defects information, the actual defect assessments on railway bogie were conducted with different types of railway bogies, which were used for the current operation. In summary, it was found that the novel infrared thermography technique could be an effective way for the inspection and the detection of surface defects on bogies since the infrared thermography method provided rapid and non-contact investigation of railway bogies.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.67 no.5
• /
• pp.669-674
• /
• 2018

Data augmentation is an efficient way to reduce overfitting on models and to improve a performance supplementing extra data for training. It is more important in deep learning based industrial machine vision. Because deep learning requires huge scale of learning data to learn a model, but acquisition of data can be limited in most of industrial applications. A very generic method for augmenting image data is to perform geometric transformations, such as cropping, rotating, translating and adjusting brightness of the image. The effectiveness of data augmentation in image classification has been reported, but it is rare in defect inspections. We explore and compare various basic augmenting operations for the metal surface defects. The experiments were executed for various types of defects and different CNN networks and analysed for performance improvements by the data augmentations.

• 전자공학회논문지 IE
• /
• v.45 no.2
• /
• pp.20-27
• /
• 2008

In this paper, we propose a new image processing technique that reliably detects the various defects of TFT-LCD polarizing films. The image of polarizing film is acquisited from reflected laser beam First, we apply the morphological image processing technique to remove the background noise. Next, we use the 2-dimensional LMS adaptive filtering and statistical characteristics to detect the white and black defects. Performance of the proposed method is evaluated on real TFT-LCD polarizing film samples.

• Proceedings of the IEEK Conference
• /
• 2009.05a
• /
• pp.237-239
• /
• 2009

The purpose of this research is to propose a system to detect a strip defect on a iron plate using an image processing, one way of finding defects on an iron plate. An existing way of image processing is using a light source which release a light energy in a certain frequency and a light absorbing display which responds to the light source. This research attempts to detect defects by using the image processing which handles an illumination, without depending on characteristics of light frequency. One of the advantages of this method is that it makes up for the weakness of the existing method which was too difficult for users to notice a defect. Also this method makes it possible to realize a real-time monitoring on a plate of iron.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.35 no.6
• /
• pp.398-406
• /
• 2015

This paper explored the results of experimental investigation on carbon fiber reinforced polymer (CFRP) composite sample with thermal wave technique. The thermal wave technique combines the advantages of both conventional thermal wave measurement and thermography using a commercial Infrared camera. The sample comprises the artificial inclusions of foreign material to simulate defects of different shape and size at different depths. Lock-in thermography is employed for the detection of defects. The temperature field of the front surface of sample was observed and analysed at several excitation frequencies ranging from 0.562 Hz down to 0.032 Hz. Four-point methodology was applied to extract the amplitude and phase of thermal wave's harmonic component. The phase images are analyzed to find qualitative and quantitative information about the defects.

• Transactions of the Korean Society of Machine Tool Engineers
• /
• v.15 no.2
• /
• pp.87-92
• /
• 2006

This paper supplements shortcoming of radioactivity check by detecting defect of SWP weld zone using ultrasonic wave. Manufacture 2 stage robot detection systems that can follow weld bead of SWP by method to detect weld defects of SWP that shape of weld bead is complex for this as quantitative. Also, through signal processing ultrasonic wave defect signal system of GUI environment that can grasp easily existence availability of defect because do videotex compose. Ultrasonic wave signal of weld defects develops artificial intelligence style sightseeing system to enhance pattern recognition of weld defects and the classification rate using neural net. Classification of weld defects that do fan Planar defect and that do volume defect of by classify.